all AI news
Medusa: Universal Feature Learning via Attentional Multitasking. (arXiv:2204.05698v1 [cs.LG])
April 13, 2022, 1:11 a.m. | Jaime Spencer, Richard Bowden, Simon Hadfield
cs.LG updates on arXiv.org arxiv.org
Recent approaches to multi-task learning (MTL) have focused on modelling
connections between tasks at the decoder level. This leads to a tight coupling
between tasks, which need retraining if a new task is inserted or removed. We
argue that MTL is a stepping stone towards universal feature learning (UFL),
which is the ability to learn generic features that can be applied to new tasks
without retraining.
We propose Medusa to realize this goal, designing task heads with dual
attention mechanisms. …
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
AI Focused Biochemistry Postdoctoral Fellow
@ Lawrence Berkeley National Lab | Berkeley, CA
Senior Data Engineer
@ Displate | Warsaw
Senior Backend Eng for the Cloud Team - Yehud or Haifa
@ Vayyar | Yehud, Center District, Israel
Business Applications Administrator (Google Workspace)
@ Allegro | Poznań, Poland
Backend Development Technical Lead (Demand Solutions) (f/m/d)
@ adjoe | Hamburg, Germany
Front-end Engineer
@ Cognite | Bengaluru